Thar's Gold in Them Thar Hills: Benefits, Cautions, and Pitfalls for Training Novices to Use Large Archival Databases

• Activity Number: 335
• Type: Topic Contributed
• Date/Time: Wednesday, August 11, 2004 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Education
• Author(s): Susan C. Losh*+ and Christopher M. Tavani*+ and Carolyn Funk*+ and Matthew DeBell*+ and Valerie Martin Conley*+
• Companies: Florida State University and Florida State University and Virginia Commonwealth University and American Institutes for Research and Ohio University
• Keywords: data archives ; large databases ; statistics training ; data analysis

Abstract

Rather than collecting new data, many scholars now employ original analyses of archival data. The increasing proliferation of (particularly online) databases brings new riches to data analysis. However, analysts often neglect to explore--or may not know--the limitations, errors, and other pitfalls in secondary data analysis before they begin. We aim our experience in different sources of error and cautions to heed toward statistics and research methods educators training advanced undergraduates, graduate students, and practitioners in the social and behavioral sciences (including educational research). Our presenters are archive creators and analysts, and primary data collectors and designers. We discuss survey archives, national educational assessments (e.g., National Assessment of Educational Progress), and post-secondary institutional characteristics (e.g., SEStat or WebCASPAR). We hope more educators will use these resources, alerting their students to the benefits and training them to ask the fruitful questions to avoid the pitfalls in their use.